McKinsey: Looking at the potential value of artificial intelligence from a national perspective

The long-term potential of artificial intelligence to transform key aspects of the way we live and support the operations of businesses, governments, and other organizations is difficult to grasp. But even today, existing and proven AI applications have the potential to create value for economies and societies around the world.

In fact, artificial intelligence has contributed to improving the quality of life at all levels of society through innovations such as predictive healthcare, adaptive education and optimized crisis response. For example, the National Health Service in the UK has established a national COVID-19 chest imaging database, which contains a shared library of chest X-rays, CT scans and MRI images to support testing and development of treatments for COVID-19 and a variety of other health conditions. status of AI technology.

Additionally, businesses are already seeing improvements in productivity and operational efficiency through initiatives such as the use of autonomous robotics in manufacturing, AI-optimized supply chains, and smart freight routing by self-driving vehicles. For example, many logistics companies are using AI-powered sorting robots to optimize their warehouse operations. Governments can also harness the power of AI by personalizing services and automating processes. Take Singapore’s “Ask Jamie,” a virtual assistant that helps citizens and businesses navigate government services across some 70 government agencies through AI chat and voice.

But governments face many obstacles – including a lack of specialist talent, limited investment in AI research and innovation, and efforts to ensure AI is used across all sectors in an ethical, safe, transparent and human-centered manner of often unclear regulations – which may prevent them from adopting AI use cases and capturing the value of AI. Indeed, when developing and deploying AI use cases, governments must proactively consider and address the rapidly changing privacy landscape and the security risks and ethical pitfalls to which AI technology may expose them.

Below, we share three steps to measure the potential impact of AI in a country and examine how this will play out in a handful of countries. We also review a range of initiatives that can help governments overcome current challenges and capture this value across the economy. These include initiating programs in areas where AI is likely to have the greatest national impact, creating a vibrant AI ecosystem, and considering appointing an AI agency, as some countries have already done.

Measuring the potential impact of artificial intelligence on a country

Artificial intelligence can mean different things to different people, but here, we use the term "artificial intelligence" to refer to data A subset of the analysis. Artificial intelligence systems include computer vision, natural language processing, and advanced robotics. They are both autonomous (perform complex tasks without human supervision) and adaptive (improve by "learning" from more data).

In September 2018, McKinsey Global Institute modeled trends in AI adoption, using early adopters and their performance as leading indicators of how full-scale enterprises may want to incorporate AI. Overall, excluding competitive effects and transition costs, early evidence suggests that AI could bring about $13 trillion in additional global economic output by 2030, increasing global GDP by about 1.2% per year.

However, while it is generally accepted that AI has great potential, it is difficult to measure and track the impact of AI in specific countries in a replicable and structured way.

Here is one approach that can help governments estimate this potential:

Identify the most relevant use case domains. McKinsey research shows that hundreds of AI use cases with the potential to unlock value across countries can be grouped into 15 areas that can inform efforts to measure impact (Exhibit 1). While each of these use case areas encompasses a number of AI applications, the underlying technologies and mechanisms by which they create value are similar, as is their typical impact when applied to similar organizations.

It may be useful to determine which use case domains are most relevant to businesses and organizations across economic sectors. For example, the areas most relevant to manufacturing are yield, energy and throughput (where AI can enable predictive machinery maintenance and resource productivity maximization) and integrated supply chain optimization (including demand forecasting and inventory optimization).

#Estimate the impact on the use case domain. Using benchmarks, the average financial impact of relevant areas can be determined at the enterprise level or at the organizational level for each priority sector. For example, if predictive maintenance in manufacturing reduced maintenance costs by an average of 10%, a company's EBIT would increase by 2%.

Expand the impact on the industry and economic levels. Suppose that for a manufacturing company, the sum of all relevant use case areas can increase EBIT by 10%. If EBIT for all companies in the industry totaled $10 billion, that EBIT would rise to $11 billion. If net operating surplus (approximately EBIT) accounts for 50% of gross value added (GVA) (i.e. contribution to GDP) in manufacturing ($20 billion in total), then AI will increase industry-wide by 10% EBIT will increase the industry's GVA by 5%. Once the sectoral-level impacts are understood, they can be scaled up to the economy-wide level by adding the impacts across all sectors.

Of course, governments must carefully consider and address the potential risks of implementing AI technologies. These risks include the following:

Privacy, Is customer privacy protected through compliance with local and global data privacy regulations?

Security, Artificial Intelligence Does the model comprehensively protect cybersecurity vulnerabilities and risks?

Fairness Is the AI ​​model fair and equitable to all customer segments?

Transparency and explainability Safety, Can you explain how the AI ​​model works and how it is used?

Safety and Performance, Is the AI ​​model fully tested to ensure it delivers every time? Required security and performance?

Third-party risks, Do all third-party vendors and partners adhere to the required risk mitigation and governance standards?

A Practical Approach: The Potential Impact of Artificial Intelligence on Gulf Cooperation Council Countries

Many countries in the Gulf Cooperation Council (GCC)—including Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates (UAE)— — is working to diversify its economy away from oil, modernize and improve efficiency through technology. Artificial intelligence is one of the technologies that many GCC countries are betting on. For example, in October 2017, the UAE government launched the UAE Artificial Intelligence Strategy, one of the first of its kind in the region. Likewise, in October 2020, Saudi Arabia’s Saudi Data and Artificial Intelligence Authority launched its National Strategy for Data and Artificial Intelligence, which aims to elevate the Kingdom to a global leader in an elite alliance of data-driven economies.

When applied to all GCC countries, McKinsey’s methodology reveals a potential value of $150 billion across all sectors of their combined economies. McKinsey analysis suggests that AI could add value equivalent to 6% or more of the GDP of each economic sector in GCC countries.

In fact, across the GCC countries, the methodology suggests that AI can play a transformative role in the public sector and manufacturing, with impact potential of 12% and 15% respectively (as a percentage of GDP) . Given that oil and gas is the largest sector in most of these economies, our analysis also reveals significant potential value; however, as the world moves away from fossil fuels, investing in new AI technologies may also be an opportunity to Accelerate necessary diversification away from oil and gas and modernize other high-potential industries.

This impact comes from different use case areas, depending on the industry. For example, in manufacturing, predictive maintenance may lead to significant productivity gains, thereby reducing downtime due to breakdowns and maintenance itself, thereby increasing output at lower costs. Advanced robotics can also play an important role in this area and significant value can be captured through data-driven supply chain optimization – for example, improving product availability, inventory costs and overall production costs through better demand forecasting.

In the public sector, successful implementation of AI could have a huge impact on countries’ wider economies. Some use cases in the industry could have a meaningful direct impact on government cash flow, such as advanced analytics for detecting fraud and incorrect payments or tax evasion in grants and transfer systems – often a significant source of financial leakage. Additionally, many public sector use cases impact populations and the overall economy by improving the quality and outcomes of public services. By providing personalized, predictive and preventive services in areas as diverse as education, transport planning and fire protection, better results may generate economic multipliers: not only can they directly save governments money, but they can also improve the lives of citizens and private companies. productive forces.

There are significant nuances across countries (Exhibit 2). For example, the methodology suggests that the potential impact of AI on the public sector may be particularly pronounced in the UAE (23%), Oman (15%) and Qatar (15%). In these countries, public service personalization, fraud and debt analysis may represent a significant portion of this potential value. Procurement and spending analytics could also be a high-opportunity use case for the public sector in Qatar and the UAE, as procurement and investment, especially in the construction sector, account for a large portion of public spending in these countries.

Meanwhile, McKinsey’s analysis suggests that manufacturing is likely to see a significant impact from AI in Bahrain (32%) and Kuwait (30%). In Bahrain, this opportunity may stem from the fact that manufacturing accounts for 18% of the economy, compared with an average of just 9% in the other five countries.

Of course, capturing this value in the GCC economies comes with potential risks with these technologies. For example, AI models used to allocate social benefits must be thoroughly examined for inherent biases against social class. Likewise, AI models used for predictive maintenance in manufacturing plants must be tested to ensure that it delivers the required performance and safety every time without any glitches.

Governments can play an active and critical role in reaping the many benefits of implementing AI technologies in a country.

Governments may have an opportunity to help seize this potential

Governments can play an active and critical role in capturing the many benefits of implementing AI technologies in a country. They can choose from a variety of approaches based on factors such as local political and economic structures. Unique challenges may arise when these approaches are applied to different sectors or countries. Based on our benchmark analysis of early adopters of AI technology, McKinsey identified several example initiatives.

Create a vibrant AI ecosystem

Countries can consider creating an AI sector or ecosystem composed of skilled practitioners, research institutions, start-ups, and large enterprises. Five enablers help create a vibrant AI ecosystem that has the potential to deliver benefits to citizens, businesses, and government entities: AI regulations; a highly skilled workforce; globally recognized cutting-edge research and innovation; A combination of domestic and foreign funding; and world-class data and computer infrastructure.

Of course, these enablers require planning and resources to achieve and are not guaranteed to have an impact. Globally, governments have taken the lead in terms of enablers:

Regulations that governments may consider given the rapidly evolving security, privacy and ethical risks associated with the use of AI. Establishing a systematic approach to proactively focus and potentially punish organizations that fail to effectively identify and manage risks associated with AI, in 2021 the EU became the first government agency in the world to publish comprehensive draft regulations specifically targeting the development and use of AI . These regulations demonstrate that organizations should have robust processes in place to manage AI risks and comply with existing and future regulations. Rather than scaling back AI development, organizations may need to create a risk management and compliance framework that allows them to continue to innovate and deploy AI at a safe pace. Additionally, governments may wish to define regulations governing citizen data based on the highest international data privacy standards, following guidance from entities such as the United Nations.

laborforce, in order to cultivate talents broadly among the population, Finland launched Elements of AI in 2018, a free online course designed to introduce non-professionals among the public AI basics. With the launch of the curriculum, the Finnish government committed to educating at least 1% of the population – a target that was quickly met and exceeded. The course is now available in more than 20 languages ​​and has been completed by 750,000 people worldwide. Similarly, AI Singapore launched LearnAI, an initiative that provides customized training courses for all segments of society, including students, educators, industry professionals and the public at large.

Research and Innovation, the U.S. National Science Foundation has established multiple national artificial intelligence institutes to focus on various artificial intelligence topics, and the U.S. government has announced an allocation of $1 billion to create 12 new AI and quantum information research and development institutes. Similarly, AI Singapore's AI Research Grants provide funding of up to S$1 million (approximately US$718,000) to researchers focusing on advanced AI to promote high-quality research aimed at developing fundamental new AI technologies, algorithms and related techniques Work.

Funding, UK Research and Innovation has committed £530 million in funding for AI R&D, including £129 million for new AI algorithms, tools and technologies, and £401 million For applications and impacts of artificial intelligence. AI Singapore's flagship program 100 Experiments (100E) aims to address the needs of different industry entities by providing funding of approximately SGD 250,000 (approximately US$180,000) per 100E project and providing access to a team of researchers and technical experts. AI problem statement.

Infrastructure, Singapore has partnered with Nanyang Technological University to establish the Center of Excellence for Autonomous Vehicle Testing and Research (CETRAN). At CETRAN, emerging AI innovations such as self-driving cars can be tested in a sandbox environment designed to replicate different elements of Singapore’s roads through common traffic plans, road infrastructure and traffic regulations. Similarly, in Taiwan, the economic ministry recently launched a self-driving car technology innovation experiment program, which authorizes a company to conduct a one-year experiment for the electric self-driving minibus Winbus.

Launching Transformative National Artificial Intelligence Plans

Some countries have also launched transformative AI plans, which they see as having the greatest opportunities for disruption based on their natural advantages. Singapore, for example, is rolling out five projects in the country's high-potential areas, including smart city and smart city transportation initiatives, a digital platform for businesses to access government services and resources, and an e-payments project.

To help determine which programs to launch, countries can consider each program’s potential economic impact, how the program aligns with strategic priorities and inherent strengths, the readiness of target sectors to adopt AI, and the program’s potential impact on citizens Influence.

Consider establishing an AI authority

Many countries, including Saudi Arabia, Singapore, the United Arab Emirates, and the United Kingdom, have established centralized AI authorities to realize the value of AI. The agency is typically responsible for overseeing the country's approach to and benefits from AI adoption, ranging from missions to mobilize the entire ecosystem to initiatives that drive the country's overall AI agenda. For example, the authority in Singapore is part of the National Research Foundation and reports directly to the Prime Minister's Office.

Governments can build the foundation for access to AI by considering some of the steps above to educate the private sector about the potential of AI, identify where the greatest opportunities lie, and support the ethical and safe adoption of AI technology. All values ​​pave the way. ways to address these technical risks. But the government cannot do this alone. For example, research institutions and universities may lead the testing and development of new AI algorithms and commercialize their solutions. They can also work with government entities to support the development of a comprehensive set of policies and guidance for organizations to ensure the safe, responsible and ethical use of AI in the technologies they develop and deploy.

Citizens can be a catalyst for change by actively demanding higher quality, faster and more personalized services from public and private sector entities; they can also be outspoken about privacy, security and privacy related to AI. and moral hazard. The greatest potential impact may come from private companies that can adopt and even innovate AI use cases, build employee capabilities, double down on AI, and adopt a systematic approach to identifying and managing AI-related risks.

In fact, working together across the economy to build innovative AI solutions and support widespread adoption of AI through skills development, investment, adoption incentives and regulations may help countries capture AI economic value. If carefully considered and unique challenges taken into account, AI may have the power to fundamentally transform societies around the world.

The above is the detailed content of McKinsey: Looking at the potential value of artificial intelligence from a national perspective. For more information, please follow other related articles on the PHP Chinese website!